Back to directory

quackd

A command-line tool that connects to multiple robots at once, gives each one a large language model brain and a smaller decision model, and lets a user state a goal from a terminal or a chat to one robot or a coordinated group, called a flock.

quackd

quackd registers each robot once by name, and a shared contract then decides which of a given robot's skills its model may use for a stated goal, how many steps it gets, and when it must ask the user first, whether the goal comes from a terminal command or a chat conversation with an assistant such as Claude. Tasks are stored as duck files and a named group of robots is a flock, so a goal can be sent to one robot or to several at once.

It supports Claude, OpenAI, Gemini, Grok, or a local model through Ollama or vLLM as the reasoning brain, plus a separate, smaller decision model for multiple-choice steps, and drives seven robot platforms today, including Microduck, Open Duck Mini, LeRobot, XLeRobot, AlohaMini, ToddlerBot, or any ROS 2 base, with one of them running on real hardware and the rest in simulation or as a mock.

More in Agent frameworks

See all 902
  • Recurse

    recurse.run

    A serverless harness that lets a coding agent design, test, and deploy its own specialist sub-agents as tools, MCP servers, or bots, defined by a schema-validated manifest.

  • ProofForge

    shulgin.is-a.dev

    An AI-agent pipeline that decomposes a mathematics problem, proves each piece, and formalizes the result in Lean 4 and Mathlib until the Lean kernel accepts it, with six of its proofs already merged into Google DeepMind's formal-conjectures repository.

  • local-coder

    github.com

    Builds a local coding-agent team on top of Ollama and OpenCode by detecting the machine's hardware, recommending models for seven specialist roles, and testing whether each assigned model can actually use tools, edit a repository and pass verification.